C86-1145 |
finding the ` uniqueness point ' for
|
word recognition
|
. Let us assume ( after Marslen
|
C86-1140 |
utterances requires more than mere
|
word recognition
|
. It is based on a number of
|
C82-1006 |
measured the goodness of each
|
word recognition
|
. To this purpose a method to
|
C02-2002 |
effect of lexicalization on new
|
word recognition
|
. We parsed all the 121,863 sentences
|
C86-1145 |
the widespread Cohort Model of
|
word recognition
|
. The parser implicit in the
|
C86-1140 |
variabilities and vagueness of the
|
word recognition
|
process , semantics in a speech
|
C80-1070 |
connected words . But in either case ,
|
word recognition
|
has been thought a basic problem
|
C86-1145 |
for the next hypothesis as to
|
word recognition
|
. This is actually enough as
|
A94-1033 |
Table 1 . The topl correct rate of
|
word recognition
|
is as low as 57 % . Relaxation
|
A94-1031 |
Table 1 . The topl correct rate of
|
word recognition
|
is as low as 57 % . Relaxation
|
C00-2169 |
Hem-nan , 1997 ) . He redefines the
|
word recognition
|
problem to identify the best
|
C86-1140 |
a satisfying certainty by the
|
word recognition
|
module . This is a very hard
|
A94-1033 |
a useful tool to post-process
|
word recognition
|
results ( -LSB- l , 4 -RSB- )
|
A94-1031 |
a useful tool to post-process
|
word recognition
|
results ( -LSB- l , 4 -RSB- )
|
acl-2001-inv1 |
with small vocabulary isolated
|
word recognition
|
and with speaker-dependent (
|
C86-1140 |
mentioned verb ) as also for the
|
word recognition
|
of the system . The special problem
|
C02-1148 |
segmenters that achieve 90 % and 95 %
|
word recognition
|
accuracy respectively . Finally
|
A92-1016 |
and surface representations .
|
Word recognition
|
is reduced to finding valid lexical
|
C86-1140 |
structures , and to guide the
|
word recognition
|
process by expectations resulting
|
acl-2001-inv1 |
acoustic and language models .
|
Word recognition
|
is carried out in one or more
|